MaLPi Intro
Published:
MaLPi (Machine Learning Pi)
First, the hardware. This is my current setup and although I’ve tested each piece separately, I haven’t had them all working together, yet.
- Raspberry Pi, model B v7 (or 0xf, I’m not sure how to read /proc/cpuinfo)
- 3D printed case (sorry, I can’t remember where my friend got the design)
- Logitech HD Webcam C525
- Vakoss USB 2.0 7 port hub (powered)
- PowerGen Mobile Juice Pack 9000
- Edimax Wireless 802.11b/g/n Nano USB Adapter
- A temporary Lego-based car body
I’m still working out how to power both the Pi and the USB hub from the battery. And I’m not sure if it’s a good idea to have the battery plugged in and charging at the same time that it’s powering the Pi.
Long Range Goal
My main goal is to have a mobile robot that uses machine learning techniques to map and navigate its way around the world. Hopefully I can find lots of intermediate steps along the way. The final test that I’m planning on using for success is if/when MaLPi can find, navigate to and plug into power when it needs to recharge it’s battery.
Next steps
- Get everything hooked up to and powered from the PowerGen battery.
- Get everything installed in a prototype wheeled rover.
- Some method of detecting low battery so the Pi can shut itself down. There are several projects out there that describe how to do that using external circuits read by the Pi’s GPIO pins. Alternatively, I may try to have the camera pointed at the battery’s status LEDs and figure it out that way.
- Manually move the rover around, taking pictures along the way, so I can get started on the software side of localization and mapping.
- Add support for several motors: wheels, steering, camera “head”.
